摘要
图像在采集和传输过程中,由于多种因素的干扰,图像会受到一定的损伤,导致图像质量下降。针对目前图像修复方法存在修复时间长,修复效果有限等不足,提出基于图像分块和改进Criminisi算法的图像修复方法。首先将原始图像划分为多个子块,然后采用Criminisi算法对每一子块图像进行修复,并针对传统Criminisi算法存在的局限性进行相应的改进,最后通过具体的图像修复实验分析所提方法的有效性和优越性。结果表明,所提方法减少了图像修复时间,可以加快图像修复的速度,图像修复后的质量要远优于对比方法,是一种速度快、效果好的图像修复方法。
In the process of image acquisition and transmission,the image may be damaged to a certain extent due to the interference of many factors,resulting in the degradation of image quality.There are some shortcomings in the current image inpainting methods,such as long inpaiting time and limited repair effect.Therefore,an image inpainting method based on image segmentation and improved Criminisi algorithm is proposed.In the method,the original image is divided into several sub⁃blocks,and then each image sub⁃block is renovated with the Criminisi algorithm.The corresponding timing is changed for the limitations of the traditional Criminisi algorithm.Finauy,the effectiveness and superiority of the proposed method are analyzed by specific image inpainting experiments.The results show that the proposed method can shorten the duration of image restoration,accelerate the speed of image restoration,and the quality of image restoration is much better than that of contrast method.It is a fast and effective image restoration method.
作者
齐巨慧
QI Juhui(Taiyuan University of Technology,Taiyuan 030024,China)
出处
《现代电子技术》
北大核心
2020年第1期63-66,共4页
Modern Electronics Technique
关键词
图像修复
图像子块
Criminisi算法
修复效率
修复效果
实验分析
image inpainting
image sub⁃block
Criminisi algorithm
restoration efficiency
restoration effect
experimen⁃tal analysis
